Analyze Data
Chi-square Example
Some research reports just list the proportion of respondents who gave each response to a question. For example, results for a question about whether survey respondents would consider purchasing a new product might look like this:
- Would consider purchasing: 40%
- Would not consider purchasing: 50%
- Not sure: 10%
While this is useful information, it’s likely to be even more useful to know whether that proportion varies based on things such as demographics, psychographics, or past purchase behavior. Chi-square tests, used in conjunction with cross-tabs, can determine this. For example, running a cross tab, we might see that most of those who would consider purchasing the new product are already customers of the company considering launching it, and a chi-square test can be used to verify that there is a statistically significant difference between customers and non-customers:
| Customers (n=500) |
Non-Customers (n=500) |
Total (n=1000) |
|
|---|---|---|---|
| Would consider purchasing | 60% | 20% | 40% |
| Would not consider purchasing | 35% | 65% | 50% |
| Not sure | 5% | 15% | 10% |
In an example such as this, the proportion of purchases coming from new as opposed to existing customers might make the difference between the product being financially viable or not. Identification of significant demographic or psychographic differences in purchase likelihoods could influence decisions such as where or how to advertise the new product if it is launched.